Uses of aporiai in Aristotle’s Generation of Animals

Jessica Gelber
(Penultimate draft – Please use published version if citing)

Given what Aristotle says about aporiai in Metaphysics B, one might expect them to
play an important methodological role in his biology.1 In Metaphysics B, we learn that going
through aporiai enables us to identify the ‘knots’ binding us, to know the direction in which
we should proceed, to recognize it when we have found what we are seeking, and to be
competent judges of competing accounts.2 These achievements would seem, at any rate, to
be equally necessary for both first- and second-order causal inquiries. Moreover, one might
expect not only for there to be something significant in Aristotle’s use of aporiai in biology,
but also for the aporiai in this empirically grounded domain to share some distinctive
features. One might suppose, for example, that the puzzles would be generated by a lack of
observable evidence as opposed to being due to some conceptual difficulty.3
It is difficult, however, to draw any general conclusions about Aristotle’s use of
aporiai in the biological treatises. Although these treatises contain many instances of aporia
language—i.e., uses of the noun and its verbal cognate—there seems to be no single purpose
for which Aristotle raises them, no common structure, and no shared source of puzzlement.
In what follows I will attempt to demonstrate this negative thesis by looking closely at the

1
Michael Boylan (“The Digestive and ‘Circulatory’ Systems in Aristotle’s Biology,” Journal of
the History of Biology, vol. 15, no. 1, Spring 1982, 89-118 at 118) goes so far as to claim that
Aristotle’s biological methodology in Generation of Animals is based on the procedure of
moving from problems to their solutions, and indeed that the biological treatises are
organized around problems. According to Gareth Matthews, Aristotle “normalizes
perplexity in the practice of philosophy” and extends the scope of this practice to deal with
“puzzles in such sciences as cosmology and biology.” (Gareth Matthews, Socratic Perpelexity,
Oxford University Press, 1999, 118-9.)
2
See Buddensiek’s essay in this volume (Chapter 7).
3
Terrence Irwin, for instance, claims that for “empirical puzzles” the “source of our puzzles
is empirical ignorance leaving us at a loss to say what happens or why it happens” (Aristotle’s
First Principles, Oxford University Press, 1988, 32). Irwin also claims that in Aristotle’s
“empirical inquiries—those in which he surveys empirical appearances—Aristotle
characteristically raises these empirical puzzles” (ibid. 42). If Generation of Animals is
considered by Irwin to be an empirical inquiry, then he is an example of a scholar who
thinks the aporiai in that treatise have a distinctive character, namely, that they share a
common source.

1
use of aporia terminology in Generation of Animals. For ease of presentation, I will follow
Aristotle in referring to what he calls an aporia by that name. One upshot of this discussion,
however, is that caution may be warranted in assuming that there is any single kind of puzzle
or difficulty being picked out by the use aporia language.

For what purpose is an aporia raised?

If one thought that the raising and resolving of aporiai were an integral feature of
Aristotle’s methodology in biology, a natural way to yoke them together might be by
reference to the purpose for which they are raised. Unfortunately, as I will show in this
section, there is no one purpose for which they are raised.
(i) In some cases, Aristotle states that some phenomenon “has much aporia” where
that announcement introduces a survey of existing views that are all equally problematic.
Such “puzzles” appear to be raised in order to establish the superiority of Aristotle’s own
theory, by showing that it avoids problems that confront the alternatives. These aporiai
perform what we might call a ‘refutative’ function.
A prime example of a refutative aporia is Aristotle’s discussion of bee generation in
GA III.10. Aristotle’s own tentatively proposed account 4 is that bees generate without
copulating, just as certain fish do.5 This is the only possible explanation, he thinks, given
what has been observed. He demonstrates this by going through the problems with all of the
prevailing alternative accounts:
Bees cannot gather their young from outside the hive, as some say. 6 If they did, these
young bees would have to be either (a) spontaneously generated, or (b) produced by another
kind of animal. Aristotle says that it is absurd, however, to think they are spontaneously

4
The proposal is only tentative, because Aristotle believes that this explanation should be
rejected if new facts are discovered, since “one must believe in perception more than
arguments (logôn)” (GA III.10, 760b31-2). The observable facts about bee generation,
Aristotle says, have not been sufficiently grasped. For a discussion of Aristotle’s
argumentation in this chapter of GA, see Joseph Karbowski, “Empirical Eulogos
Argumentation in GA III.10,” British Journal for the History of Philosophy, 22, 1, 2014, 25-38.
5
The way bees generate is unique, despite its similarity to how it goes for some fish. The
kinds of fish that generate with copulating produce other fish of the same kind, whereas only
the king bee produces other kings (760a4-9).
6
Aristotle also notes that since no bee larvae are gathered from outside the hive, it is also not
possible for only some to be generated and others gathered.

2
generated. For, he reasons, in this case we should have found spontaneously generated bees
that are not collected remaining in the places where other bees would have collected them.
However, this is not something we observe. Aristotle also rejects (b), the idea that bees could
collect offspring that are produced by another kind of animal. For, animals generate
offspring that are the same in kind as themselves, and so this would violate a principle that
holds of animals, generally.7 Finally, against both (a) and (b), he notes that while it makes
perfect sense for bees to collect honey, since honey is their food, it is absurd to think that
bees would collect anything that is not food.
It cannot be that bees are produced by the copulation of drones and worker bees, as
others claim, either. Generation by copulation requires both male and female principles. So
drones and worker bees would have to exhibit sexual differentiation. But it does not seem as
though worker bees can be either male or female: Worker bees cannot be female, because
they have stingers, and nature does not assign weapons to females. Worker bees cannot be
male, since worker bees tend to the young, and in Aristotle’s view, males never do that.8 In
fact, generation by copulation seems to be impossible not only on the assumption that two
different kinds of bees, such as workers and drones, copulate with each other to generate all
of the kinds, but also on the assumption that the members of the same kinds of bees
copulate with each other to produce another individual like themselves. This is apparent,
Aristotle claims, since young drones are produced when no drone is present, and worker
bees are only produced when king bees are also around.
The only viable option, Aristotle concludes, is that worker bees are hermaphroditic
and generate drones without copulation, and that king bees generate both worker bees and
kings. Although it is a bizarre theory, it is the only one that makes sense of all of the
empirical evidence and coheres with his zoological principles. Consequently, Aristotle thinks
it is the best theory, given what we have observed.

7
Aristotle knows about hybrids such as donkeys, however, although he does not
acknowledge such exceptions here.
8
For discussions of Aristotle’s use of such principles having to do with males and females,
e.g., that nature does not give weapons to females and that males do not tend to their
offspring, see G.E.R. Lloyd, Science, Folklore, and Ideology: Studies in the Life Sciences in Ancient
Greece, Cambridge University Press, 1983, 94-105.

3
 (ii) At other times, an aporia is presented not for the purpose of rejecting other
theories, but rather because doing so will advance Aristotle’s own inquiry in some way.
These aporiai move the investigation forward by making clear that there is a genuine problem
or question needing to be addressed, and by making explicit what the constraints on any
adequate theory or explanation are. I will label this a ‘zetetic’ use of aporia.

(iia) In some cases, Aristotle advances the inquiry by motivating the introduction of
something unobservable that would make sense of the phenomenon being discussed. One
such motivating, zetetic aporia is raised in order to justify positing the presence of pneuma in
the male’s sperma.9 At GA II.2, 735a29-736a23, he says that someone might be puzzled10
about the nature of sperma, since it does not behave in the way that either watery or earthy
substances do. Unlike watery substances, it does not thicken (pachunetai) when cooled. Unlike
earthy substances, it does not thicken when boiled, as milk does. But according to Meteorology
IV.4, the moist and the dry (referred to as ‘watery’ and ‘earthy’ in the GA passage) are the
passive potentials out of which all uniform bodies are made.11 So it seems that it must be one
or the other.
Aristotle begins his resolution of this aporia by asking: “or have we not distinguished
all of the things that occur (sumbainonta)?” (735b7-8). He points out that in addition to watery
and earthy moistures, there are also those made up of water and pneuma, for example foam,
oil, and lead-ore. So, he now introduces a further feature of his account: Semen is a

9
Sperma in this discussion clearly refers to the male semen, although Aristotle will sometimes
use this term to refer to the female spermatic residue, as well as the mixture of male and
female residues—the “seed”.
10
It is not entirely clear whether the expression aporeseien an tis means that “someone might
be puzzled” or that “someone might raise a puzzle.” The same is true about the use of the
noun aporia; it is not clear whether aporia refers to a state of puzzlement or the difficulty that
gives rise to puzzlement. Consequently, it is not as obvious to me as it is to Garreth
Matthews that for Aristotle “an aporia is not a state of puzzlement at all; it is rather a puzzle
or conundrum or difficulty that produces, or is likely to produce, a state of puzzlement.”
(“The Normalization of Perplexity in Aristotle,” in May Sim (ed.) From Puzzles to Principles?:
Essays on Aristotle’s Dialectic, Lexington Books, 199, 125-136 at 130)
11
“We must discuss the forms of the passive potencies, the moist and the dry. The passive
principles of bodies are moist and dry, whereas other things are combinations from these,
and a body in its nature consists more of whichever of the two is more—for instance, some
things consist more of dry, others of moist.” (Meteor. IV.4, 381b23-5)

4
compound of pneuma and water, and pneuma is warm air.12 This accounts for the properties of
sperma, and more importantly, the presence of pneuma is now a feature of the embryological
account that he can appeal to in explaining other phenomena.13

(iib) Another type of zetetic use of aporia occurs when Aristotle wants to make some
corrections or revisions. Some of these zetetic aporiai are raised in order to correct what has
been (wrongly) taken to be the empirical facts. He says, for instance, that someone might be
puzzled about the growth of eggs (GA III.2, 752a24ff). Eggs do not have an umbilical cord,
he says, so they do not receive nourishment in the way that animals that are born live do.
They also do not simply use themselves as nourishment in the way that larvae do.14 How,
then, does the nourishment get inside the egg in the first place, when that egg is hard and
lacks anything like the umbilical cord?
Aristotle resolves this by wheeling in some empirical data that has been overlooked
(lanthanei): at the outset the egg is soft, though it quickly hardens once released. If it were not
soft, egg-laying would be painful. That initial, soft egg has a very small umbilical-like ‘pipe’
protruding from the pointed end, by which it was attached to the uterus. This is difficult to
see, but Aristotle says it has been observed in aborted chicken eggs.15

(iic) Other zetetic, corrective aporiai are raised when qualifications or modifications to
the explanation that Aristotle has offered need to be made. For instance, in the course of
explaining what happens just after fertilization in GA II.4, Aristotle says that while all of the
parts are “potentially” present in the kuêma, which is the first mixture of male and female

12
For the same reasons, Aristotle calls the nature of olive oil aporôtata: if it were mostly water,
it should be solidified (pêgnusthai) by cold, and if it were mostly earthy, it should be solidified
by fire (Meteorology IV.7, 383b20ff). As it is, neither heat nor cold solidifies it, and both
thicken it. His solution to the puzzle is to point out that oil is full of air (aer). For a
discussion of the “olive oil aporia”, see James Lennox, “Aristotle on the Emergence of
Material Complexity: Meteorology IV and Aristotle’s Biology,” HOPOS: The Journal of the
International Society for the History and Philosophy of Science, vol. 4, Fall 2014, 272-305 at
288ff.
13
See Peck’s discussion in his Loeb edition of GA, Appendix B for an overview of the role
of pneuma in Aristotle’s accounts of movement, nutritive soul activities and sensation.
14
As Aristotle explained in GA II.1, 732a32, it is only a part of the egg that becomes the
living organism, and the rest is used as the nourishment for growth. See also the discussion
in GA III.9.
15
cf. HA 559b16

5
spermatic residues,16 the first part to be actually present is the heart. The heart has to be
present first, because it is the source (archê) of all the other parts, both homoeomerous and
heterogenous. The new fetus, while still incomplete, and still only potentially an animal,
needs nourishment from elsewhere. Thus the heart sends off blood vessels (the umbilical
cord) to the uterus to procure the nourishment, just as seeds in the ground send off roots
and shoots. But about this, Aristotle says that someone might be puzzled.

“Someone might aporêseie, if on the one hand the blood is nourishment, but
on the other hand the heart—being bloody—comes to be first, but
nourishment [comes from] outside (thurathen), 17 from where does the first
nourishment enter?”(740b2-5)

Aristotle has just said that the growing fetus gets its nutriment from the uterus, to which the
heart sends vessels. Now he points out that this cannot be the way the heart gets its initial
nourishment. So how does the heart receive it? To answer this, he makes a qualification to
the theory:

“Or is this not true, that all [nourishment comes from] outside, but rather
just as in the seeds of plants there is some such thing that first appears milky,
so also in the matter of animals there is residue for the assemblage
straightway.”(740b5-8)

So, there is something not quite correct after all about the idea that nourishment comes from
outside the heart.

(iid) The previous two examples of zetetic aporiai are ones that allow Aristotle to
make a correction, either to his own theory or to what is taken to be the empirical data about

16
“Well then, among those animals not having separate males and females, for these the
sperma is like the kuêma. I mean by kuêma the first mixture from the female and the male.”
GA I.20, 728b32-4
17
The idea that nourishment must come from something outside is not explicitly mentioned
by Aristotle in GA. As far as I know, the closest he comes to saying this is the claim (at
724b34) that nourishment (trophê) is manifestly epeisakton.

6
some phenomenon. A further purpose for which zetetic aporiai arise is to motivate not
merely a correction but rather what we might think of as a radical change in focus. An
extended discussion in GA II.1 is a good illustration of the type of use I have in mind.
Aristotle’s purpose in raising this aporia, as I understand it, is to prime his audience for a
surprising shift in the way we should conceive of agency—of what it is to be an agent of
some change—by painstakingly running through the difficulties that will arise if such a shift
is not made.18
At 733b32, Aristotle raises a question about what the agent of embryonic
development can be. Here he is concerned to identify the agent that produces an embryo’s
body parts out of the first mixture of male and female spermatic residues, a mixture that he
sometimes refers to as the kuêma and sometimes the sperma. This agent must be either
something external or internal to this mixture, but there does not appear to be any viable
candidate. First, Aristotle rules out the first possibility:

“For either something external or something present in the semen and the
sperma makes [the parts], and the latter is either a certain part of soul or soul,
or might be something having soul. Well, surely that something external
makes each of the viscera or other parts would seem absurd. For it is neither
possible for something not touching to move [anything], nor possible for
something to be acted upon by something not moving [it].” (GA II.1,
733b32-734a4)

According to Aristotle’s general account of change, there must be an agent with an active
potential to impose form and a patient with a passive potential to receive form. But that is
not sufficient: conditions must obtain such that the change occurs, namely, that the agent
and patient be in contact. It is this general principle—that change requires contact between
the agent and patient—to which Aristotle refers here.19 In his theory of animal generation,

18
For a similar use of aporia, i.e., one that motivates a conceptual shift, see de Caelo II.12,
292a19-22: “We may object that we have been thinking of the stars as mere bodies, and as
units with a serial order indeed but entirely inanimate; but should rather conceive them as
enjoying life and action. On this view the facts cease to appear surprising.” (Stocks trans., my
emphasis)
19
Aristotle discusses contact in Generation and Corruption I.6-7 and Physics III.2.

7
the male parent has the active potential to impose form, but he never makes direct contact
with the matter that has the passive potential to receive that form. So, it seems that there
must be some part already present in the kuêma that is the agent, since only something in the
kuêma could make contact with the matter. However, this option is also ruled out.
Aristotle’s lengthy rejection of this second option involves establishing that (1) if the
agent were some part present in the kuêma, it would have to be a part of the embryo that is
being formed.20 But (2) since there is no part of the body that does not have soul, some
ensouled part would have to be present from the outset (734a14-16). Yet this, too, is
impossible: (3) no part of the embryo can be present in the seed, because seed is simply what
the parts come to be out of, and so must be produced prior to the parts that come to be out
of it (734b1-2). 21 So, nothing can be made simultaneously with seed such that it is both
present in the seed from the outset and also comes to be out of seed.

Aristotle resolves this by showing how we might reconceive of the agent in a way
that makes it possible for it to be something both internal and external, after all.

20
He does so by setting up a dilemma: If the agent were something existing in the kuêma, it
would have to be either a part of it, or something separate from it (734a5-6). The second
horn, i.e. that the agent be something separate, is unreasonable (alogon) (734a5-7). For, if that
were the case, once the animal has been generated, then that separate part either remains or
perishes (734a7-8). It is evident that nothing remains in the new animal or plant that is not
also a part of it (734a8-9). So, it would have to be something that perishes, which is also
absurd (atopon) (734a10). Why? If it perished, it would have to do so either after having made
all of the parts or after making some of them (734a9-10). If it perishes after making some of
them, then how are the other parts formed (734a10-11)? Aristotle’s reasoning seems to be
that if we said that this separate part makes some of the internal organs, e.g. the heart and
liver, and then perishes, it is unclear how the rest of the parts and organs are formed. If we
say that it is the heart and liver that forms the rest of the parts, there is nothing to prevent
these from also perishing after they make the next organ. By parity of reasoning (according
to the “same logos”), the parts that make the other parts would also perish, but they survive
(734a11-13). Thus, the agent would have to be a part in the kuêma, and not something
separate (734a13-14).
21
I have omitted one epicycle in this long argument: The parts are produced in succession,
since the idea that they come to be simultaneously is ruled out “by perception” (734a20-1).
Parts formed earlier cannot be agents of later parts, because that conflicts with another
general principle, namely, that the agent must have the form in actuality that which is being
generated (734a29-31). The idea that the heart, for example, has the form of the liver is
absurd (734a31-3).

8
 “We must try to resolve these. For perhaps something we said is not
unqualified,22 namely how in the world it is not possible to come to be by
something external. For in one way it is possible, but in another it is not.”
(GA II.1, 734b4-7)

There is no agreement among interpreters about how to understand the discussion that
immediately follows. 23 As I understand it, the resolution involves showing that we were
wrong to be looking for some thing—“some this” (tode ti, 734b18)—that could be making the
embryo’s parts. It is rather the movement being conveyed in the semen that is making the
parts.24

“Surely, to say “sperma” or “what sperma comes from” makes no difference,

insofar as it has in it the movement which that one [sc. the father] was
moving. And it is possible for this to move this, and this to move this, and to
be like the spontaneous marvels. For in a way the parts, although resting,
retain a power. Whenever something external moves the first of the parts,
straightaway the following one comes to be in actuality. Just as, then, in the
spontaneous [cases], [a] in one way [the mover] moves, not touching
anything now, but having touched. And similarly, also what sperma comes
from or the one who made the sperma did touch something, but is no longer
touching. [b] And in another way the internal movement [moves], just as the

22
This is the fallacy of secundum quid: “confusing what is general (haplôs) and what is not
general but some particular”, e.g., that ‘non-being’ exists, since what-is-not is what-is-not
(Rhetoric II.24, 1402a3-17). See also SE 166b37ff.
23
See Henry 2005 for a discussion the interpretations offered by Alexander and Simplicius,
as well as one of his own. This is one of two passages in GA which draws on an analogy
with the spontaneous marvels (automata tôn thaumatôn), and it is unclear what kind of
spontaneous marvels Aristotle has in mind. It is clear enough that the father is being
compared to the “first external mover” that gets the marvel moving, but beyond that the
details are all disputed. I will ignore this controversy here, since the point this example is
meant to illustrate has to do with the purpose for which Aristotle raises this aporia. I think
that everyone agrees that the point is to provide a model for conceiving of the agent as
something both external and internal.
24
Johannes Morsink (Aristotle on the Generation of Animals: A philosophical study, University Press
of America, 1982, 98) also takes the point of the automata analogy to be to “teach us not to
view the efficient cause as a thing or a tode ti but as a movement (kinêsis – b17) or power
(dunamis – b12).”

9
 house-building [makes] the house. Well then, that there is something which
makes, not being some particular present in [the sperma], nor from the
beginning completed, is clear.” (GA II.1, 734b7-19)

We need not worry that the father is no longer in contact with the semen and directly
moving it, for it is really the movement that the father had set up (by concoction of the
spermatic residues) that is doing the work. And that movement in the semen can continue
after the father releases the semen, much like the “spontaneous marvels” can continue to
move after an external mover moves a part of it. So, since the movement in the semen is
making the parts, there is a sense in which (tropon de tina, 734b16) the agent is something
internal; since the father set up at that movement, there is a sense in which (tropon men tina,
734b13) the agent is something external.

(iii) Finally, sometimes Aristotle claims there is an aporia where it is not clear that
what he goes on to say is moving the inquiry along in any way at all. In fact, in at least some
cases, it appears that he is simply pointing out that there is some phenomenon for which we
lack any explanation. These uses seem to be merely a way of introducing the next topic to be
discussed. An example of this sort of use is the aporia about the uterine mole (mulê):

“There is a puzzle, on account of what [uterine moles] have not come about
among other animals, unless something has entirely escaped notice.” (776a8-
9)

The uterine mole is an abnormality that occurs early on during pregnancy (a “molar”
pregnancy) whereby a mass is formed inside the uterus. Aristotle claims that the cause of this
phenomenon is a weakness of the heat relative to the matter that the heat needs to concoct.25
But apparently, Aristotle reports, this phenomenon has only been observed to occur in
human women, and not in other kinds of animals. So, there is a question, introduced by the

25
Modern biologists attribute this phenomenon to an over-production of the tissues that are
supposed to become the placenta. Interestingly, since the placenta is what feeds the fetus,
Aristotle was not far off (even if only by accident) in thinking that the cause of uterine moles
has to do with excess nourishment.

10
aporia terminology, about why this should be the case. Perhaps there is something unique
about human females that would explain this:

“One must suppose (dei nomizei) that the cause is that the women alone of
other animals is husterikon and excessive in evacuations (katharseis) and not
able to concoct them. So whenever a kuêma is put together from moisture
that is difficult to concoct, then the so-called mulê comes about, reasonably
either especially or uniquely in women.” (776a9-14)

Although it is not entirely certain what it means for women to be husterikon, 26 it is this
feature, specific to human women, that accounts for the mola uteri only occurring in
humans.27

From this survey of a variety of passages, it appears as though there are a number of
purposes for which Aristotle raises what he calls an aporia. Some are raised in order to refute
alternative theories (refutative aporiai) by showing that they involve some absurdity or
conflict with the observed facts or zoological principles, while others advance Aristotle’s
own project (zetetic aporiai). Moreover, the zetetic aporiai move the inquiry forward in several
ways: they either offer occasion to posit some unobservable entity, or to add new empirical
data, or to revise the details of Aristotle’s own theory, or to change the conceptual
framework being employed. Finally, claiming that something has or involves an aporia
sometimes seems merely to be Aristotle’s way of introducing the next topic for discussion,

26
It does not seem as though husterikon means merely “suffering in the womb,” “hysterical,”
or “of or belonging to the womb,” which are the meanings listed in LSJ (online version,
August 2015). It likely has rather to do with overt menstruation, which is most apparent in
humans. Other references to husterikon in GA include husterikous membranes (717a5);
husterikon part (morion) (720b21); husterikon place (738b7). At HA 570a5, (cf. 566a11), the
husterikous passages are contrasted with seminal passages (thorikous).
27
A case might be made for thinking that Aristotle’s purpose in raising this aporia about the
uterine mole is to advance his inquiry, since it allows him to point to a ‘fact’ about human
women not previously mentioned. Thus one might prefer to consider his purpose here to be
zetetic. I am proposing, however, that this use of aporia be placed in a distinct category.
Unlike the introduction of pneuma or revisions to Aristotle theory of the sort we have seen,
the claim that human women alone are husterikon does not play any further role in Aristotle
theory, as far as I can tell.

11
where he seems to mean that someone might not know how to answer a question and thus
be as yet “without passage”.

Structure

In our previous discussion, we have seen that there are a variety of purposes for
which Aristotle raises aporiai. Now, we will see that an aporia can take a number of different
forms.
(i) Some of the aporiai are presented in the form of a dilemma. In such cases, we are
given what appear to be exhaustive options. In some of these, the options are equally
unappealing or impossible (the “negative” aporiai). The discussion of the agent of embryonic
development is an example of a negative aporia. Recall that it initially seemed as though the
agent must either be something internal to the kuêma or something external to it, and it did
not seem that either could be the case. Aristotle resolved this by showing how, in a way, the
agent can be both: Since it is movement—not “some this”—that is the agent, and that
movement is derived from the male parent and conveyed to the kuêma, it is both something
external (the father’s movement) and something internal (the movement in the first mixture
of spermatic residues). Aporiai of this type have the form “Is A or B the case?”

(ii) In other cases, we are given what looks to be an exhaustive and exclusive
dilemma, but no considerations for or against one of the horns are given. For example, in an
aporia about what occurs at “the beginning of generation” in GA V.1, Aristotle begins by
providing two options.

“There is a puzzle concerning the beginning of generation: does wakefulness

obtain prior, or sleep?” (GA V.1, 778b23-25)

Without even considering the possibility that wakefulness is the prior state, Aristotle
proceeds to adduce reasons in favor of it being sleep that is prior. It would be reasonable to
think sleep is the earlier stage, he says, since animals become more wakeful as age advances,
suggesting that they are proceeding towards wakefulness from its opposite. Moreover, since
animal generation is a transition from not-being to being, it is plausible to think that an

12
animal would first be asleep, since that appears to be intermediate and “like a boundary
(methorion) between living and not living” (778b29-30). However, Aristotle is reluctant to call
whatever the developing animal does at the beginning “sleep”. Wakefulness is or at least
involves the exercise of sense perception, and its opposite—sleep—is the absence sense
perception. But it is not simply any absence of perception that constitutes sleep, but only the
absence of perception in creatures for which perception can be present. Plants do not sleep,
and they cannot wake up from being asleep, and that is because plants do not perceive.28 The
state plants are in is something else. So, too, at the earliest stage of development when the
animal lacks the ability to perceive, we must say that the state they are in is not exactly sleep,
but something like sleep.

“If it is necessary for an animal to have perception, and whenever

perception has first come to be it is at that time first an animal, then it must
be supposed that the condition from the beginning is not sleep but like
sleep, just as the sort that plants also have.” (GA V.1, 778b32-35)

Here, the aporia is initially about whether A or B is the case, but Aristotle
immediately moves to consider only B. This looks less like a genuine dilemma than a
question about whether and how B might be the correct answer to the question that was
originally posed.

(iii) Furthermore, some aporiai do not have the appearance of a dilemma at all. Many
times Aristotle says that there is an aporia about some phenomenon, where what he
apparently means is that there is a question about it, such as how, what or why something
happens.

(iiia) Some of these questions are of the form “Why does F occur among A’s but not
among B’s?” Typically, these are questions about why something generally true of some
larger kind is not true of some subset of that kind. He uses the aporia language, for instance,
28
cf. de Somno I, 454b25-29: “We say that sleep is in a certain way a motionlessness (akinêsia)
of perception, and like a restraint (desmos), and that the release and relief is wakefulness. But
no plant is able to partake in either of these affections; for without perception, neither sleep
nor wakefulness obtain.”

13
when he introduces the discussion of a difference between bird and fish generation: bird
eggs are separated from the uterus prior to being ‘complete’, whereas fish eggs remain
attached to the uterus. Aristotle says that “someone might aporêseien about why the
generation of birds and fish differ in this respect” (754b20-1). Both are egg-layers, but the
particular ways in which they produce eggs is not the same. 29 So, Aristotle sometimes
introduces with the aporia language a question about why something true of A is not true of
B, where A and B are members of some wider kind, such as egg-layers or live-bearers.30

(iiib) Aristotle also uses the aporia language where what follows is simply a question
about what happens. For example, he begins GA II.3 by asking what happens to the bodily
part of semen when it enters the uterus, given that that bodily part does not become part of
the developing embryo (736a24ff). And sometimes these are questions about why something
happens. He says that the reason why the eyes appear largest at the beginning of generation
but are the last of the parts to be completed “has” aporia (743b32ff). What Aristotle says
involves aporia here is simply a question about the cause of some phenomenon that we do
not yet know.

There does not appear to be any significant overlap among the aporiai if we attend to
the form in which they are presented, i.e., whether as dilemmas or merely questions. So, just
as we were not able to draw any general conclusion about the purposes of Aristotle’s
biological aporiai, there is also no such conclusion to draw about their structure.31

29
Besides the fact that birds and fish are both egg-layers, it is not clear why someone would
expect bird and fish generation to be the same. Aristotle has discussed many other
differences between birds and fish earlier in GA. And Aristotle does not spend any time
explaining why this difference, in particular, should be puzzling. Rather, he immediately
proceeds to simply give the explanation.
30
Another aporia having this structure concerns a peculiarity of selachia (757a14ff). Female
selachia are not seen discharging their kuêmata, and the males are not observed emitting their
milt. Both of these phenomenon, however, are seen in other live-bearers. Here the question
being raised is why selachia do not exhibit all of the characteristics of other live-bearers.
31
Is there some other feature of their structure that might lead us to some general
conclusion? One might wonder, for example, whether there is some important difference
that is being tracked by occurrences of aporia versus aporeseien an tis. Unfortunately, however,
this seems not to be the case. For, Aristotle uses both the noun and verbal forms when
discussing the same phenomenon on several occasions. Both aporia and aporeseien an tis are
used in the discussions of, for example, the behavior of semen (735a29ff), why the female

14
Sources

Having been unable to discover any common purpose for which Aristotle raises
aporiai in biology, and having seen that they lack any common form, one might wonder if
instead there is some other distinctive feature they share. Perhaps, one might think, the
biological aporiai are united in being generated by having a common type of source, such as a
lack of observational evidence. However, there does not seem to be anything general to
conclude about the reasons—what I am calling the “sources”—why the aporiai arise.

(i) Some involve a tension between Aristotle’s biological theory, as it has been
articulated so far, and some apparently recalcitrant data or observations. The tension that
gives rise to the aporia about the heart’s nourishment (740b2-8) is of this sort. The theory
established the heart as the source of nourishment for the developing embryo, but the heart,
too, has to get its nourishment from somewhere. Here the conflict is resolved by simply
modifying the theory: Perhaps there is nourishment in the heart from the outside, so it is not
true that all nourishment comes from somewhere else.

(ii) Other aporiai arise because there is a conflict between the details of Aristotle’s
biological theory and his own deeply held, general commitments. Attending to these is often
illuminating, for they reveal a natural scientist working within the constraints of a grand
metaphysical framework, self-consciously applying the concepts and principles he argues for
elsewhere, though at a fairly abstract level. As I understand it, the aporia about the agent of
embryonic development has this type of source. Aristotle has argued elsewhere that in
general, change requires contact between agent and patient. Here the aporia arises because it
seemed difficult to see how his embryological theory could accommodate this.

(iii) Further, there are cases said to involve aporia in which there is a tension between
some general principles and the empirical evidence. For example, Aristotle announces at GA
I.8, 718a35-7 that “someone might be puzzled about the facts concerning the uteri in

needs the male in order to generate (741a6ff), and whether the cause of multiple births and
redundant parts is the same (770b30ff).

15
women, how they are” since “many oppositions (hupenantiôseis) belong to them”(718a36-7).
In some kinds, the uterus is higher up, near the diaphragm, and in some it is lower, near the
pudenda.
Five chapters earlier, (GA I.3, 716b12-13) Aristotle had already said that “the things
concerning” (ta peri) the testes and uteri are not the same in all blooded animals, and then
proceeds to discuss differences amongst the male sexual organs for the next five chapters.
He explains why some kinds have and some lack testes, and why those which have them do
so internally or externally. Thus someone might rightly be puzzled about why variations
amongst the female sexual organs should be so puzzling.32
It appears as though this aporia arises because the variation in the position and
orientation of the uterus does not coincide with the distinction between being a live-bearer
and egg-layer. Some kinds of live-bearing animals have the uterus high, and others have it
low. The same is true of the egg-laying creatures. However, the variations found among male
testes, too, cuts across the live-bearer/egg-layer distinction, and that is not something
Aristotle identified as puzzling. Why might someone be puzzled about the variations in the
position of uteri in different animal kinds, but not about the variations in the placement of
testes?
Although Aristotle does not say so explicitly, I suspect that the reason this is
puzzling is that it is in tension with some familiar principles of his natural science, such as
the principle that nature does nothing in vain, and that morphê follows function (and not the
reverse). In the case of the uterus, there seem to be variations in morphê that do not track
variations in its function, which is to provide “protection, shelter, and concoction” (I.12,
719a33-4). Given this function, one might expect that only a difference in what is being
produced (i.e., whether what is being protected, sheltered, and concocted is an egg or a
‘complete’ animal) should require a difference in the position of the uterus. But not all egg-
layers have the uterus positioned in the same way. The function of the testes is to “make the
movement of the spermatic residue more steady” (717a30-1). Given that function, there is
no reason to expect the male organ’s presence or position to co-vary with the kind’s being an

32
This seems to be reflected in, for example, Platt’s translation: “It is not easy to state the
facts about the uterus in female animals, for there are many points of difference.” Platt
seems disinclined to treat this as a puzzle at all, but rather a “difficulty” in “stating facts”
about the uterus.

16
egg-layer or live-bearer. This is why there is puzzlement about the uterus but not about the
male sex organs.
Aristotle resolves this by distinguishing different ways of being egg-layers, and
different ways of being live-bearers. Egg-layers are divided into the kinds that produce
“complete” eggs (i.e., hard-shelled ones, such as those of birds), and those that produce
“incomplete” eggs (i.e., soft ones, such as fish produce) that are “completed” externally. For
the hard shell to be formed, the uterus needs to be near a source of heat, and this is what the
higher placement provides. If less heat is required, as is the case for the formation of the
soft-shelled, incomplete eggs, the uterus is positioned lower and closer to the exit (peras),
since that is more expedient. So, there is a difference in function from which the differences
in morphê follow, after all.

(iv) Finally, sometimes the source of aporia is not any difficulty or tension of the sort
we have seen in the previous examples. Sometimes the reason that there is aporia is that some
fact has been overlooked. For example, it has “escaped notice” that bird eggs have an
umbilical cord-like part at the beginning, and that missing observation is what gives rise to
the difficulty in explaining how the egg receives its nourishment. And sometimes it is not
that we lack observations, but it is simply the fact that no explanation has yet been offered
that creates the impasse.

So, difficulties can be due to conflicts between some particular explanation, whether
Aristotle’s or someone else’s, and the empirical evidence. An aporia can also involve tensions
between some particular explanation and some more general principles. Some of the
principles are specific zoological ones, such as the principles that nature does not give
weapons to females, or that males do not tend to their offspring. Other principles are ones
that hold of change, generally, such as that change requires contact. Further, sometimes the
source of aporia is that the observed data do not cohere with the general principles. Lastly,
for some of the aporia, it is not obvious that there is any genuine conflict or tension at all.
These appear to be cases where the aporia is simply due to an absence of evidence or lack of
any explanation, at least so far. Consequently, aporia in Aristotle’s biology cannot be said to
be generated by some specific type of problem. There is no obvious way, at any rate, to unify
the aporiai in GA by their source.

17
Conclusion

From a close look at the discussions in which Aristotle explicitly uses the aporia
terminology, it appears as though aporia in GA is pollakôs legomenon. The purposes for which
Aristotle raises them vary a great deal, as do the structures the discussions exhibit, as well as
their sources. Moreover, it is possible to point to further ways in which they differ. For
example, Aristotle uses a variety of methods to resolve them. Sometimes he brings in
additional observations, such as the fact that eggs do have something umbilical cord-like,
even though it is hard to see.33 At other times, he modifies or qualifies his theory in some
way, as he does when discussing the source of the heart’s nourishment. More interesting
cases involve his making some conceptual distinction, as he does when trying to identify the
agent that makes the embryo’s body parts.
The aporiai also differ quite a bit in terms of their seriousness. Some, to be sure, are
deeply problematic, and Aristotle seems to think that their resolution is pressing. Others,
however, do not seem to be particularly worrisome to him at all, at least in the present
context. In fact, sometimes he seems merely to be pointing to a potential problem for his
theory, but one that he is not going to attend to at this time. After raising the issue about
heart’s nourishment, for instance, Aristotle returns to discussing how the growing embryo
receives its nourishment through the umbilicus, and makes no further mention of the heart’s
nourishment.34

Now, what can we conclude about Aristotle’s biological aporiai, given that they seem
to be so multiform and multipurpose? One might reasonably harbor doubts that those
puzzles that he calls aporiai mark off some well-delimited group. Perhaps the concept of an
aporia may be more fluid for Aristotle than one might have supposed. It may be simply a
matter of being without passage, in some vaguely defined sense.

33
See also, e.g., his criticism of the idea that hyenas have two pudenda (757a2-13). “Cursory”
or “casual” (ek parodou) observation has produced this false belief.
34
The same is true of the aporia in GA V.1 about whether sleep or wakefulness is prior. After
concluding that it is not sleep but something “like sleep” that is the prior state, he
immediately goes on to the next topic.

18
 Alternatively, one might maintain that he does in fact have some clearly defined
notion of an aporia, such as the state of puzzlement arising from dialectical problemata,35 while
thinking nevertheless that such genuine aporiai are not tracked by the use of aporia language.
For, the multiplicity of purposes, structures and sources of the aporiai we have looked at
suggests that no common kind of puzzlement is being indicated by the occurrence of the
aporia terminology. At best, the genuine aporiai would have to be merely a subset of the
difficulties Aristotle calls by that name. So, if he thinks an aporia is a specific type of
puzzlement, his use of aporia terminology will not be sufficient for telling us when we are
facing one. We also have reason to doubt that the occurrence of aporia terminology is
necessary for telling us when there is a genuine aporia: There are at least a few instances in
Generation of Animals where one finds discussions that are very similar in structure and
purpose to some of those we have examined. For instance, in his survey of the various
arguments adduced by the preformationist theory of generation36 in Book I, chapters 17-18,
there is the same systematic treatment of preformationist arguments as one finds in passages
explicitly flagged as involving aporia. This is also true of his discussion of alternative
explanations of sexual differentiation in Book IV, chapter 1. The purpose for which he
methodically runs through the arguments and rival views, just as in the aporia about bee
generation, is to show the superiority of his own theory and the impossibility of any other.
In neither case does Aristotle employ aporia language.
Finally, we might conclude that aporia is not an especially important part of
Aristotle’s method in his biology. When studying the heavens, perhaps aporiai are needed,
since “we have but little to go upon, and are placed at so great a distance from the facts in
question” (de Caelo II.12, 292a15-7, Stocks trans.). The study of eternally ungenerated and
imperishable beings is more valuable and more divine than that of generated and perishable
beings, but it is harder to study the former since “those things on the basis of which one
would examine them and those things about them which we long to know, the perceptual
phenomena are altogether few” (PA I.5, 644b25-6, Lennox trans.). Where observable data is
available, on the other hand, perhaps the knots are readily apparent, and thus running
through the puzzles is not required. In the case of the study of plants and animals,

35
As discussed by Rapp in his contribution to this volume (Chapter 6).
36
According to the preformationist theories, the seed from which the organism develops is
composed of parts from each parent’s body.

19
consequently, using aporia might thus be less pressing: in this domain, we have more means
(euporoumen mallon) to attain knowledge, since we are fellow creatures (suntrophon) (PA I.5,
644b28-9).

References

M. Boylan, “The Digestive and ‘Circulatory’ Systems in Aristotle’s Biology,” Journal of the
History of Biology, vol. 15, no. 1, Spring 1982, 89-118.

D. Henry, “Embryological Models in Ancient Philosophy,” Phronesis 50 (1), 2004, 1-42.

T. Irwin, Aristotle’s First Principles, Oxford University Press, 1988.

J. Karbowski, “Empirical Eulogos Argumentation in GA III.10,” British Journal for the History of
Philosophy, 22, 1, 2014, 25-38.

J. G. Lennox, “Aristotle on the Emergence of Material Complexity: Meteorology IV and

Aristotle’s Biology,” HOPOS: The Journal of the International Society for the History and
Philosophy of Science, vol. 4, Fall 2014, 272-305.

G. Matthews, Socratic Perpelexity, Oxford University Press, 1999.

G. Matthews, “The Normalization of Perplexity in Aristotle,” in May Sim (ed.) From Puzzles
to Principles?: Essays on Aristotle’s Dialectic, Lexington Books, 1999, 125-136.

J. Morsink, Aristotle on the Generation of Animals: A philosophical study, University Press of

America, 1982.

A.L. Peck, (trans. and comm.) Aristotle: Generation of Animals, The Loeb Classical Library,
Harvard University Press, 1942.

A. Platt, (trans.) Generation of Animals, in J. Barnes (ed.) The Complete Works of Aristotle: The

20
Revised Oxford Translation, vol. 1, Princeton University Press, 1984, 1111-1218.

J.L. Stocks, (trans.) de Caelo, in J. Barnes (ed.) The Complete Works of Aristotle: The Revised Oxford
Translation, vol. 1, Princeton University Press, 1984, 447-511.

21